a multiple-level theory leadership: impact - dr. hatfield multiple level... · 2015. 7. 24. · a...

A Multiple-Level Theory of Leadership: TheImpact of Culture as a Moderator

Kyoungsu Kim, Chonnam National UniversityFred Dansereau, State University of New York at Buffalo, Buffalo, NY

In Sook Kim, Yonsei UniversityKong Soo Kim, Kunsan National University

In an effort to present a multiple level theory ofleadership, we suggest a set of conditions thatmay moderate the acceptance of a leader’sbehavioral patterns. By considering leadershiptheories at multiple levels of analysis and byconceptualizing culture at a higher level ofanalysis, we suggest that it depends on culturewhether subordinates accept a leader’sbehavioral pattern. We consider the followingpatterns: a leader (1) displays a consistent stylethrough the use of consideration and initiatingstructure, (2) differentiates between in- and out-groups of subordinates via delegation, or (3)interacts with subordinates on a one-to-one or

dyadic basis via providing a sense of self-worth.We also discuss the implications of this

approach and suggest directions for futureresearch.

In organizational science, variousresearchers have suggested that it may be usefulto integrate different concepts or theories at

different levels of analysis (Dansereau, Alutto,& Yammarino, 1984; Roberts, Hulin, &

Rousseau, 1978; Rousseau, 1985). Nevertheless,there are very few attempts to do so. In this

paper, we attempt to show an integration of thetwo different concepts of culture and leadershipthat are typically viewed as at different levels ofanalysis.

In terms of culture many theorists andresearchers have conceptualized culture as

varying from individualism to collectivism

(Hofstede, 1980; Kagitcibasi, 1997; Kim et al.,1994; Triandis, 1994). The individualism-collectivism dimension refers to the relationshipamong individuals. Individualism is defined as a

social pattern of loosely connected individualswho emphasize independence, whereascollectivism is defined as a social pattern of

closely linked individuals who emphasizeinterdependence (Markus & Kitayama, 1991;Triandis, 1995). Therefore, along theindividualism-collectivism continuum,individuals are viewed from the individual levelof analysis where each individual is independentof higher levels of analysis (e.g., group) to theother end of the continuum where individualsare viewed as interdependent or as embedded ina group.

To show how these different views mayrelate to levels of analysis, we consideredWeick’s (1978) view of interdependence that itcomes about from individuals interacting (andforming dyads). From this perspective, webelieve that the development of interdependencecan be viewed as involving a shift from theindividual level of analysis where eachindividual is independent to a higher level ofanalysis (e.g., dyad or group) where individualsbecome interdependent. This notion of

interdependence among individuals has alsobeen applied to leaders and followers - i.e.,leadership. Many leadership theorists andresearchers have suggested that leadershipinvolves interdependence between leaders andfollowers as well as among followers (e.g., Bass,1990; Dansereau, Graen, & Haga, 1975; Graen,Liden, & Hoel, 1982; Weick, 1978; Yukl, 1998).

From the above brief consideration of

leadership and culture, it seems that both cultureand leadership focus on independence and

interdependence and accordingly on levels of

by guest on July 24, 2015jlo.sagepub.comDownloaded from

http://jlo.sagepub.com/

79

analysis. To combine these views, we addressthe question of under what cultural conditionsmight interdependence between leaders andfollowers occur and when mightinterdependence not occur. In other words, weask how, though at different levels of analysis,do culture and leadership interact? To

accomplish this objective, we consider two

conditions. In one condition, individuals are

viewed as becoming independent with a leader;we call this a lack of acceptance of the leader.

Whereas, in the other condition, individualsbecome interdependent, we call this acceptanceof the leader. We define &dquo;interdependence&dquo; as

one’s influence over another individuals and

&dquo;independence&dquo; as one’s lack of influence overanother individuals. When two individuals are

interdependent, one individual influences whatthe other individual does (e.g., Weick, 1978). Inother words, the other individual accepts oneindividual. Likewise, when two individuals areindependent, one individual has no influencewhat the other individual does. In other words,the other individual does not accept one

individual.

Acceptance and lack of acceptance are

associated with the issue of ethical concernsabout leader’s behavioral patterns. A leader’sbehavior may be value-free. In other words, aleader behaves depending on his or her ownstyle, differentiation of group members, or one-on-one exchange relationship. However, a

leader’s behavior may become ethical or

unethical depending on whether the behavior isaccepted or not. For example, a leader’sbehavior accepted by subordinates may beethical to them, whereas a leader behavior

unaccepted by subordinates may be unethical tothem. We propose that acceptance or lack ofacceptance depends on cultures. Therefore, it iscritical to investigate which behavioral pattern isaccepted (ethical) or unaccepted (unethical) indifferent cultures.

When a leader is accepted, the leaderinfluences followers. For example, it is temptingto hypothesize that if a leader acts in a friendlyor supportive manner, shows concern and

respect, or provides a sense of self-worth

support, followers may tend to be satisfied withthe leader and show higher performance.However, it has been suggested that leaderbehavior may not relate to effectiveness even if

leaders and followers are interdependent becauseof a number of other contingency factors

(Ayman, Chemers, & Fiedler, 1998) andsubstitutes for leadership (Kerr & Jermier, 1978;Podsakoff & MacKenzie, 1998). Therefore,when a leader is accepted, leader behavior mayor may not relate to effectiveness (satisfactionand performance). On the other hand, when aleader is not accepted, the leader does not

influence followers. Followers ignore the leader,do not follow the leader, and do not changeattitudes and behavior based on what the leaderdoes. Therefore, although a leader acts in a

supportive manner, shows concern, and providesa sense of self-worth support, this leaderbehavior cannot lead to effectiveness, becausethe leader and followers are independent. Thisgoes to the heart of the question of effectiveness.If a leader and followers are not interdependent,there is no leadership effect. Thus,interdependence becomes a necessary but not asufficient condition to examine leadershipeffectiveness.

To summarize, our specific purpose here isto explore what factors influence

&dquo;interdependence&dquo; and &dquo;lack of

interdependence.&dquo; By doing so, we take intoaccount when (under what conditions) followersaccept their leaders as legitimate, that is, leadersand followers become interdependent. Wedefine conditions for acceptance of the leader interms of culture. Culture has been defined as&dquo;shared values&dquo; about appropriate behaviors orbehavioral patterns (Pascal & Athos, 1981;Schein, 1992). Therefore, if a leader’sbehavioral pattern is congruent with values

shared by followers, the leader will be acceptedand leaders and followers become

interdependent. However, if not, followers willnot accept the leader and leaders and followersremain independent. Thus, culture influenceswhether followers accept leader behaviors or

behavioral patterns.We select culture as a key condition for

several reasons. First, a number of leadershiptheorists view culture as a key factor that maymoderate the relationship between leader

behavior and leadership effectiveness (Bass,1990; Dorfman, 1996, 1998; Dorfman & Ronen,1991; Hartog et al., 1999; House, 1995). Second,there is a need for establishing a theory to

explain differential leader behavior and



80

effectiveness across cultures (House, 1995:

House, et al., 1999). Third, culture is a

multidimensional construct (Singelis, Triandis,Bhawuk, & Gelfand, 1995; Triandis, 1995;Triandis & Gelfand, 1998). Therefore, a set ofmultiple-level approaches to leadership can becompared under various conditions of culture.Fourth, culture is usually conceptualized at a

higher level of analysis than leadership and thusprovides a basis for making predictions from ahigher to lower levels of analysis.

To present the culture-based conditionsunder which a leader is accepted, we first brieflyoutline a set of three multiple-level approachesto leadership - i.e., average leadership style(ALS), vertical dyad linkage (VDL), andindividualized leadership (IL). Although thereare many theories at different levels of analysis,our focus is on the interdependence that

develops between leaders and followers.

Interdependence between leaders and followerscan occur between leaders and followers wherefollowers are viewed as independent amongthem or interdependent (Dansereau et al., 1984;Dansereau et al., 1995). Second, becauseindividualism and collectivism have beenviewed as including both vertical and horizontalattributes (Triandis, 1995; Singelis et al., 1995),we conceptualize culture in terms of four typesof cultural conditions - i.e., horizontalcollectivism (HC), vertical collectivism (VC),horizontal individualism (HI) and verticalindividualism (VI). Third, we discuss how thesecultural conditions may influence whetherfollowers accept their leaders (because of

interdependent with the leader). Throughout thisdiscussion, we develop testable research

propositions.

Theoretical Background

Three Approaches to Leadership: ALS,VDL, AND IL

As noted previously, in the literature threedifferent approaches have been developed to

show interdependence between leaders andfollowers. First, according to the ALS approach,superiors (leaders) view subordinates (followers)in the same group as a whole and they do notdiscriminate among subordinates (Fiedler, 1967;Fleishman, 1957, 1998; Likert, 1961;Schriesheim, Cogliser, & Neider, 1995; Stogdill

& Coons, 1957). One approach that includes thisidea views superiors behaving toward allsubordinates by displaying a consistent stylethrough the use of consideration and initiatingstructure. Therefore, subordinates as a group arehomogeneously influenced by such a consistentstyle and link with superiors.

Second, Dansereau et al. (1975) developedthe VDL approach as an alternative to the ALSapproach, which was later called the LMX

approach (Graen et al., 1982; Graen & Scandura,1987; Graen & Uhl-bien, 1995). Because theLMX approach is rather ambiguous in terms oflevels of analysis (Schriesheim, Castro, &

Cogliser, 1999), we will focus here only on theVDL approach. According to the VDL

approach, superiors develop differentiated

relationships with subordinates in the same

group on the basis of loyalty, attraction,contribution to the group, quality of leader-member exchange, or some combination of thesefactors (Dansereau et al., 1975; Dienesch &

Liden, 1986; Liden & Maslyn, 1998;Schriesheim et al., 1999; Schriesheim, Castro, &

Yammarino, 2000). That is, superiors establish aspecial relationship with their in-groupmembers, whereas they establish a formal

relationship with their out-group members.Based on these two different relationships,superiors influence in- and out-group-membersin two different ways - i.e., leadership andsupervision (Dansereau et al., 1975), or moreprecisely delegation and lack of delegation,respectively (Schriesheim, Neider, & Scandura,1998). Unidirectional downward influence, a

formal employment contract, and the lack ofnegotiating latitude characterize supervision orlack of delegation. On the other hand, leadershipor delegation is characterized by delegation ofsupervisory activities that go beyond the formalcontract (Dansereau et al., 1975; Dinesch &

Liden, 1986; Graen & Uhl-Bien, 1995;Scandura, 1999; Schriesheim et al., 1998;Wayne, Shore, & Liden, 1997). Therefore,subordinates in the same group are

heterogeneously influenced by the leader’sdifferentiation of in- and out-group.

Third, Dansereau (1995) and his colleagues(Dansereau et al., 1995) developed the IL

approach, which they show is quite differentfrom the ALS and VDL approaches (Mumford,1998; Mumford, Dansereau, & Yammarino,



81

2000; Schriesheim et al., 1999). According tothe IL approach, superiors view theirsubordinates as independent individuals ratherthan as a group and interact with subordinates ona one-to-one or dyadic basis. &dquo;A leader’s one-to-one or dyad-based treatment&dquo; can be

distinguished from &dquo;a leader’s differentialtreatment.&dquo; On the one hand, a leader’sdifferential treatment is based on social

comparison in which a leader comparessubordinates on the basis of their loyalty,attraction, or contribution to the group and

distinguishes in-group members from out-groupmembers. In this case, it depends on the othermembers whether a subordinate becomes in-

group or out-group member. Therefore, the samesubordinate can be an in-group member in onework group, but can be an out-group member inanother group, depending on the other membersin a work group. On the other hand, a leader’sone-to-one or individualized treatment is basedon social exchange (Homans, 1961). In this case,

Figure 1 shows three different approaches of leadership: (1) ALS, (2) VDL, and (3) IL.

As Figure 1 shows, superior-subordinaterelationships can be viewed as reducible to (1)whole person and whole work-grouprelationships, as in the ALS approach, (2)superior parts and work group pars relationship,as in the VDL approach, and (3) superior-subordinate dyads, as in the IL approach. In theALS approach, a superior as a whole personhomogeneously influences subordinates in thesame group - i.e., whole work groups. However,in the VDL approach, a superior differentiatesamong subordinates in the same group (i.e., in-and out-group members) and differentiallyinfluences subordinates. In other words, a

superior is divided into two parts: one deals with

leaders and subordinates form independentdyads (Dansereau et al., 1995; Schriesheim et

al., 1999). Therefore, if a leader gets muchbenefit from a subordinate, he or she will givemuch benefit in return. The IL approachassumes that (1) the particular subordinate withwhom a superior interacts influences thebehavior that a superior displays toward thatsubordinate and that (2) a superior’s support forself-worth is reciprocated by a subordinate’s

providing satisfying performance on a dyadicbasis. (Dansereau et al. [1995] defined self-worth as a judgment or perception that a leadersupports a subordinate’s self-worth by (1)paying attention to the subordinate’s individual-level needs and feelings, (2) assuring that theleader has confidence in subordinate’s

motivation, integrity, and ability, and (3)supporting the actions a subordinate takes.)Therefore, each subordinate in the same group isinfluenced by the leader’s self-worth support ona one-to-one or individualized basis.

in-group members and the other deals with out-group members. In the IL approach, a superiorneeds to be neither whole person nor parts of aperson. In other words, a superior behavestoward a subordinate depending on how he orshe behaves toward the superior - i.e., a superiorand a subordinate have one-on-one relationshipindependent of the group.

A superior may behave toward subordinatesin the same group on the basis of (1) his or herown style, (2) differentiation of in- and out-

group, or (3) one-to-one or dyad. As notedpreviously, differential treatment is based on

social comparison in which a leader comparessubordinates on the basis of their loyalty,



82

attraction, or contribution to the group.However, one-to-one or dyadic treatment isbased on social exchange in which a leader and asubordinate form an independent dyad. Thus, aleader may show different behaviors (self-worthsupport) toward subordinates in the same groupdepending on each subordinate’s performance.However, a leader may also show high degree ofself-worth support toward subordinates in thesame group, when subordinates provide thesame satisfying performance. Thus, these twotreatments are not based on the leader’s own

style. If a leader has a certain style (e.g., task-oriented), he or she tends to show the samebehavior (i.e., task-oriented behavior) toward allsubordinates in the group, regardless of whethersubordinates differ in their ability or

performance.It may help to distinguish two cases: (1) the

case where the performance of subordinates issimilar to each other and the leader treats

subordinates on a group basis and (2) the casewhere the performance of subordinates is similarto each other and the leader treats subordinateson a dyadic basis. In the former case, a leaderdisplays the same behavior toward subordinateson the basis of his or her own style. If a leader istask-oriented, he or she displays task-orientedbehaviors toward all the subordinates regardlessof their performance level. However, in the lattercase, a leader may show the same behaviorstoward a set of subordinates, becausesubordinates in the same group show the same

performance (returns). In other groups, a leadermay display different behaviors, becausesubordinates in the same group show different

performance (returns). In yet other groups a

leader may display differential behaviorsbecause subordinates in the same group showdifferential performance (returns). That is, in theindividualized case, leader behavior does not

depend on the leader’s style and not on a

differential basis, but rather on eachsubordinate’s performance alone (returns).

According to Dansereau et al. (1984), a

variable is viewed as including entities or levels.Therefore, leader behavior (a variable) mayshow group homogeneity, heterogeneity, or

independence (levels). A leader may displayconsideration behavior on the basis of (1) his orher own style or personality characteristics, (2)differentiated relationships with subordinates, or

(3) a one-to-one or dyadic basis. However,according to more traditional approaches to

leadership (i.e., ALS approach), a leader’sconsideration behavior is based mainly on theleader’s people-oriented style (Fleishman, 1998;Stogdill & Coons, 1957). Therefore, a leader’sconsideration behavior varies between persons(leaders) and thus varies between groupsbecause of group homogeneity. In a similar way,according to the VDL approach, a leader’s

leadership- and supervision-oriented behavior isbased on the differentiation between in- and out-

group. Therefore, such behavior varies withinpersons (leaders) and thus varies within groupsbecause of group heterogeneity. According tothe IL approach, a leader’s self-worth supportbehavior is based on the degree to which eachsubordinate provides satisfying performance.Therefore, a leader’s self-worth supportbehavior varies between dyads.

We will suggest in a subsequent sectionthat it depends on culture whether subordinatesaccept leader behavior (e.g., consideration,leadership-oriented, or self-worth supportbehavior) and whether leaders and subordinatesbecome interdependent around these behaviors.But first, in the next section we conceptualizeculture in terms of two dimensions - i.e.,individualism-collectivism and vertical-horizontal.

Conceptualization of CultureIndividualism-collectivism has been viewed

as a key factor in defining culture (Hofstede,1980; Kagitcibasi, 1997; Kim et al., 1994;Triandis, 1994, 1995). For example, Triandisviews individualism-collectivism as &dquo;the most

important world view that differentiatescultures&dquo; (1994: 286). Nevertheless, it has been

argued that the individualism-collectivismdimension is too broad and thus fails to

differentiate countries in the same culture (Chen,Meindl, & Hunt, 1997; Kagitcibasi & Berry,1989; Schwartz, 1990; Singelis et al, 1995;Triandis, 1995). Triandis (1990, 1995) suggeststhat one way to resolve this problem is to

consider an attribute of individualism-collectivism -- i.e., vertical and horizontal. Heargues that both individualism and collectivism

may be horizontal and vertical, which results infour types of cultures - i.e., horizontalcollectivism (HC), vertical collectivism (VC),



83

vertical individualism (VI), and horizontalindividualism (HI).

Figure 2 summarizes the four types ofculture based on the two different dimensions ofindividualistic-collectivistic and vertical-horizontal. In a vertical-collectivistic culture,people value status differentiation and group

membership, whereas in a vertical-individualistic culture, people value status

differentiation and independent individuals. In ahorizontal-collectivistic culture, people value

equality and group membership, whereas in ahorizontal-individualistic culture, people valueequality and independence.

These four types of culture have receivedsome empirical support (e.g., Triandis and

Gelfand, 1998). Accordingly, in this paper weconceptualize culture in terms of the four types.

Acceptance Of Leader UnderDifferent Cultures

We now consider the three different waysthat a superior behaves toward subordinates -i.e., ALS, VDL, and IL - within the four

patterns from culture - i.e., HC, VC, HI, and VI.

Acceptance of Leader in HorizontalCollectivism (HC)

HC is characterized by a cultural pattern inwhich the individual sees the self as an aspect ofa group and the same as the self of others. That

is, the self is merged with the members of thegroup, all of whom are extremely similar to eachother and thus equality is the essence of thiscultural pattern. In this pattern, peopleemphasize common goals with others and

interdependence (Singelis et al., 1995: 244). InHC, such as in Japan and Israel, people believethat individuals in the same group should betreated on an equal and group membership basisregardless of whether they give more input orless output. They also believe that group

membership should determine what they receiveand how they are treated (Triandis, 1995).Therefore, a leader should treat subordinates

equally or homogeneously by displaying his orher own style.

A leader may homogeneously show high orlow degree of consideration behavior based onhis or her own style. If a leader shows a highdegree of consideration, subordinates should

accept the leader because they are

homogeneously treated. In a similar way, if aleader shows a low degree of consideration,subordinates should accept the leader because

they are homogeneously treated. Therefore,acceptance of the leader is not based on thenature of treatment (e.g., high or low

consideration), but on the distribution of

leadership (homogeneity, heterogeneity, or

dyadic exchange). In HC, people also believethat group resources must be distributed on an

equal and group membership basis (Singelis etal., 1995). Leader behaviors such as displayingconsideration can be intrinsic rather thanextrinsic resources to subordinates of the group.If a leader distributes these resources on an

equal or group membership basis, subordinatesshould tend to accept the leader’s behavioral

(distribution) pattern as legitimate because groupmembership determines what they get.



84

It is plausible that members in the samegroup may show similar or homogeneousbehaviors in the culture of HC and thus a leaderhas too little basis to distinguish members.However, subordinates in the same group maydiffer in their ability and motivational forcesbecause of individual differences. Therefore,individuals may differ in their behaviors and

performance. A leader may perceive thedifferences of ability, motivation, and

performance. Thus, a leader may treat

subordinates on a differentiation basis or on aone-on-one (dyadic) basis. However,subordinates tend not to accept suchdifferentiation or dyadic treatments as

legitimate. On the other hand, if a leader treatssubordinates equally or homogeneously in a

group, subordinates should be likely to acceptthe leader’s behavioral pattern as legitimate,because they are treated equally and

homogeneously.In terms of previous theorizing, Misumi

(1995) suggested in Japan a leader

homogeneously behaves toward subordinates inthe same group on the basis of PM (production-maintenance). The concept of PM is similar tothe concept of consideration (or people-oriented)and initiating structure (or task-oriented) (e.g.,Misumi & Peterson, 1985). Through a

randomized whole-group design in Israel, Eden(1990) suggested that Pygmalion effects becreated in whole groups without interpersonalcontrast effects in which leaders are led to

expect high performance from some

subordinates and low performance from theother subordinates. In terms of these studies, theauthors did not test whether leaders show

homogeneous behaviors or expectations towardsubordinates in the same group. Therefore, thefollowing proposition needs to be viewed as onetheoretical assertion.

Proposition 1. In horizontal collectivism,subordinates are more likely to accept the leaderbehavioral pattern based on an ALS approachinvolving consideration and structure than an ILor a VDL approach as legitimate.

Acceptance of Leader in VerticalCollectivism (VC)

VC is characterized by a cultural pattern inwhich the individual sees the self as an aspect ofa group, but the members of the group differ

from each other, some having more status thanothers do. People emphasize the importance ofstatus and hierarchy and thus inequality is theessence of this cultural pattern. (Triandis, 1996;Triandis & Gelfand, 1998). In VC, such as inKorea and China, people stress values of

inequality as well as group membership(Triandis, 1995). Accordingly, on the one hand,people believe that individuals in the same

group must be treated on a type of groupmembership basis (Triandis, 1995). On the otherhand, in VC, people stress the values of

hierarchy, rank, and prestige that are oftendetermined by age, seniority, loyalty, affection,contribution to the group, or some combinationof these factors. People believe that individualsin the same group should be treated

differentially on the basis of such factor(s).Therefore, in VC, if a leader provides a specialtreatment with high-ranking subordinates, but

provides a formal treatment with low-rankingsubordinates, subordinates would perceivefairness from the leader. Thus, subordinateswould accept the leader as legitimate, when aleader treats subordinates differentially (in- andout-group basis) in the same group.

In VC, people essentially believe that groupresources should be distributed on a groupmembership basis but also on a differentialbasis. Leadership-oriented behaviors (e.g.,delegation and negotiating latitude) can beresources to subordinates. From a differentiation

perspective, such resources should be distributedon the basis of the rank in the hierarchy. If aleader distributes his or her resources towardsubordinates equally or homogeneously in thesame group, subordinates especially those withhighly ranked in the hierarchy would view sucha behavioral pattern as unfair and would not

accept the leader as legitimate because they arenot preferentially treated. In Korea, people inpublic schools tend to accept equal-based butseniority-based pay system. In other words,equal pay should be paid to people with thesame seniority (tenure), but different pay shouldbe paid to people with different seniority.However, people tend not to accept job-based,performance-based, and skill-based pay system.Therefore, official ranking based on

performance does not exist. Chen, Meindl, andHunt (1997) suggest that in China individualswith a VC support the differential reward



85

allocation reform. This suggests that if a leaderbehaves or distributes resources on a differential

basis, subordinates in the same group would

accept the leader. However, these authors didnot test whether, in China, leader behaviors varywithin groups - i.e., whether leaders showbehaviors on a differential basis. Therefore, thefollowing proposition needs to be viewed as onetheoretical assertion.

Proposition 2. In vertical collectivism,subordinates are more likely to accept the leaderbehavioral pattern based on a VDL approachinvolving delegation than an IL or an ALS

approach as legitimate.

Acceptance of Leader in HorizontalIndividualism (HI)

HI is characterized by a cultural patternwhere an autonomous or an independent self ispostulated, but the individual is equal in statuswith others (Singelis et al., 1995: 245). In thispattern, people want to be unique and distinctfrom groups and are highly self-reliant, but theyare not interested in becoming distinguished orin having high status. In HI, such as in Swedenor Australia, people stress the values of equalityand independence. People believe that they areall equal in status and that they are unique anddistinct from groups and highly self-reliant

(Singelis et al., 1995; Triandis and Gelfand,1998). In this cultural pattern, a person needs togive something in order to get something inreturn. Therefore, when an individual receivesbenefits from another individual, he or she mustgive some benefits in return. If an individualbreaks this exchange rule, he or she would notbe accepted as an exchange partner. In a similarway, subordinates in a group believe that theirleader should treat them as an independent. If aleader treats the subordinates independently andprovides some benefits with each subordinate,the subordinates will accept the leader as

legitimate and give some benefits in return.In HI, people also believe that group

resources should be distributed on an

independent basis. Self-worth support behaviorscan be resources to subordinates. Accordingly, ifa leader shows such behaviors to each individualsubordinate depending on how much the leadergets some benefits from the subordinate, eachsubordinate perceives that he or she is treated onan independent basis. In this case, subordinates

will accept the leader as legitimate. Crouch andYetton (1988), from a study of Australian

managers and subordinates, suggested that

manager-subordinate dyads with high taskcontacts and friendly behaviors show highperformance, whereas manager-subordinatedyads with low task contacts and low level offriendliness show low performance. However,the authors did not test whether in Australia, aleader’s behavior (task contacts and friendliness)varies between dyads - i.e., whether

interdependence between leaders andsubordinates exists. Therefore, the followingproposition needs to be viewed as one

theoretical assertion.

Proposition 3. In horizontal individualism,subordinates are more likely to accept the leaderbehavioral pattern based on an IL approachinvolving support for self-worth than an ALS ora VDL approach as legitimate.

Acceptance of Leader in VerticalIndividualism (VI)

VI is characterized by a cultural pattern, inwhich individuals see the self as autonomous or

independent, but individuals see each other asdifferent, and thus inequality is the essence ofthis pattern (Singelis et al., 1995: 245). In thispattern, people want to become distinguishedand acquire status. In VI, such as in the UnitedStates, people believe that they should be treatedas being independent and that they should bedistinguished from others in status (Singelis etal., 1995). They also believe that the more theygive, the more they get (Triandis, 1995).Therefore, when an individual receives benefitsfrom another individuals, he or she must givesome benefits with the same value in return. Ifan individual breaks this exchange rule, he orshe would not be accepted as an exchangepartner. In a similar way, subordinates in suchgroups believe that a leader should treat them onan independent and equity basis. That is, a

leader should treat each subordinate

independently depending on how much eachsubordinate provides some benefits to the leader(Singelis et al., 1995).

In VI, people also believe that groupresources should be distributed on an

independent and equity basis. Self-worth supportbehaviors can be resources to subordinates.

Accordingly, if a leader shows such behaviors to



86

each individual subordinate depending on howmuch the leader gets something benefits fromthe subordinate, each subordinate perceives thathe or she is treated on an independent basis andequity basis. Therefore, subordinates will acceptthe leader as legitimate. We view the IL

approach as acceptance of the leader in

Individualism (horizontal and vertical). Ourview corresponds with Triandis’ argument, &dquo;it

should be understood that even verticalindividualistic culture are rather horizontal,because all individualistic cultures,..., are

horizontal&dquo; (1995: 46). Dansereau et al. (1995)analyzed the data from a public university,manufacturing firms, a supply firm, and a

hospital in the United States and found thatleader behavior varies only between dyads - i.e.,interdependence between leaders andsubordinates exists at the dyad level of analysis.These findings suggest that when a leader treatssubordinates on a one-to-one or dyadic basis,subordinates may accept the leader.

Proposition 4. In vertical individualism,subordinates are more likely to accept the leaderbehavioral pattern based on an IL approachinvolving support for self-worth than an ALS ora VDL approach as legitimate.

Acceptance of Leader and EffectivenessThe relationship between acceptance of

leader and outcome variables or effectiveness

(e.g., satisfaction and performance) may beclarified by considering the distribution of

leadership (homogeneity, heterogeneity, or

dyadic basis) and the nature of leadershiptogether. Subordinates in the same group can behomogeneously and poorly treated, out-groupmembers are poorly treated, and each individualcan be poorly treated depending on his or herown performance (returns). In HC, subordinatesaccept their leader, when the leader

homogeneously and poorly behaves toward thesubordinates, which may not lead to low

satisfaction, motivation, and performancebecause of a number of contingency factors

(Ayman, Chemers, & Fiedler, 1995) or

substitutes (Kerr & Jermier, 1978). For example,when subordinates are highly matured, leader’spoor behavior (e.g., low degree of considerationand structuring behavior) may lead to highsatisfaction and performance (e.g., Hersey andBlanchard, 1982).

In HC cultures, if a leader behaves

homogeneously and displays high degree of

consideration and structuring behavior, theleader will be accepted. However, subordinates(or group) may or may not show highsatisfaction and performance, because of a

number of contingent factors and substitutes. Ina similar way, in HC, if a leader behaves

homogeneously and displays low degree of

consideration and structuring behaviors, theleader will be accepted, which may or may notlead to low satisfaction and performancebecause of a number of contingent factors andsubstitutes.

In VC cultures, if a leader behaves

heterogeneously and displays high degree ofleadership-oriented behaviors toward in-groupmembers and low degree of such behaviorstoward out-group members, the leader will beaccepted. However, in-group members may ormay not show high effectiveness (satisfactionand performance) because of a number of

contingent factors and substitutes. And out-

group members may or may not show low

effectiveness, because of contingent factors andsubstitutes. In VI and HI culture, if a leaderbehaves on a dyadic basis and displays highdegree of self-worth support, the leader will beaccepted, which may or may not lead to higheffectiveness because of contingent factors andsubstitutes. In a similar way, if a leader behaveson a dyadic basis but displays low degree ofsuch behaviors, the leader will be accepted,which may or may not lead to low effectiveness.

Table 1 shows the summary of the

relationships between cultural patterns,acceptable leadership, and possible extensions ofacceptable leader behavior. As noted previously,in the culture of HC, subordinates tend to acceptthe superior who shows behaviors based on theALS approach, whereas, in the culture of VC,subordinates tend to accept the superior whobehaves based on the VDL approach. In thecultures of VI and HI, subordinates tend to

accept the superior who behaves on the basis ofthe IL approach. The three different approachesof leadership (ALS, VDL, and IL) assert

different leader behaviors. The idea is that inhorizontal-collectivism cultures the group-basedbehaviors would tend to define leadershipwhereas other types of actions such as

delegation and individual support for self worth



87

would not. Thus certain cultures would valuecertain behaviors as leadership and other

cultures would not value such behaviors as

leadership.

Table 1

Acceptable Leadership and Leader Behaviors in Different Cultural Patterns

Discussion

In this paper, we have suggested that it isuseful to integrate different concepts or theoriesat different levels of analysis. To show an

integration of different concepts or theories, weconsidered leadership theories at multiple levelsof analysis and conceptualized culture at a

higher level of analysis and suggested one waythat culture can interact with leadership. In

summary, we proposed that a leader may behavetoward subordinates in the same group on thebasis of (1) a consistent style, (2) thedifferentiation of in- and out-group, and (3) one-to-one dyad. Focusing on the notion of

interdependence (acceptance of leader) and

independence (lack of acceptance), we proposedthat leaders’ behavioral patterns may be

accepted by followers as legitimate (or leadersand followers become interdependent)depending on subordinates’ shared beliefs of thenature of leader treatment.

We focus on both leadership style and onthe distribution of leadership, because a variable(leader behavior) is viewed as including entitiesor levels (homogeneity, heterogeneity, or dyadicexchange, see Dansereau et al., 1984).According to more traditional approaches to

leaders (ALS approach), a leader displaysconsideration and structuring behaviors on thebasis of his or her own style or personality

characteristics or a homogeneous fashion. Forexample, a leader would show consideration andstructuring behaviors on the basis of his or herpeople- and task-oriented style. However,according to the VDL approach, a leader showsleadership- and supervision-oriented behaviorson the basis of in- and out-group differentiation- i.e., a heterogeneous fashion. For example, aleader would show delegating behaviors towardin-group members. According to the IL

approach, a leader shows self-worth supportbehavior on the basis of dyadic exchange. Forexample, a leader would show high degree ofself-worth support toward a subordinate who

provides satisfying performance. Therefore,although it is critical to assess which behavioralpattern (homogeneity, differentiation, or dyadicexchange) subordinates accept, a specificbehavior is based on a specific behavioral

pattern. Thus, both types of leader behavior andbehavioral patterns should be considered at thesame time.

We define acceptance of leader as

interdependence between leaders and followersand lack of acceptance as lack of

interdependence between leaders and followers.We also suggest that it depends on culturewhether a leader is accepted. For example, in

HC, when a leader homogeneously behavestoward subordinates in the same group, theleader will be accepted regardless of whether the



88

leader shows high or low degree ofconsideration and structuring behavior. In a

similar way, in VC, when a leader

heterogeneously behaves toward subordinates

(i.e., a leader shows leadership-oriented andsupervision-oriented behaviors toward in- and

out-group members, respectively), in and out-

group members may accept the leader. In HI

and VI, when a leader behaves towardsubordinates on a one-to-one or dyadic basis, theleader will be accepted regardless of whether theleader shows high or low self-worth support.

In this paper we have attempted to suggestthat it may depend on the culture and sharedvalues whether subordinates accept the leader’sbehavioral pattern as legitimate. However,acceptance of leader is not equated with

motivation, satisfaction, and performance. Therelationship between acceptance of leader andoutcome variables (e.g., satisfaction and

performance) may be clarified by consideringthe distribution of leadership (homogeneity,heterogeneity, or dyadic basis) and the nature ofleadership together. Subordinates in the same

group can be homogeneously and poorly treated,out-group members are poorly treated, and eachindividual can be poorly treated depending onhis or her own performance (returns). In HC,subordinates accept their leader, when the leaderhomogeneously and poorly behaves toward thesubordinates, which may not lead to low

satisfaction, motivation, and performancebecause of a number of contingency factors

(Ayman, Chemers, & Fiedler, 1995) or

substitutes (Kerr & Jermier, 1978). For example,when subordinates are highly matured, leader’spoor behavior (e.g., low degree of considerationand structuring behavior) may lead to highsatisfaction and performance (e.g., Hersey andBlanchard, 1982).

Some theorists of culture have argued thatculture may directly influence leader behaviorsor behavioral patterns and that leader behavior

might be consistent with culture (e.g., Erez &

Earley, 1993; Offermann & Hellmann, 1997).Nevertheless, it has been generally accepted thatnational culture includes various subcultures thatdiffer from the culture and that other factors

such as individual characteristics (Bass, 1990;Yukl, 1998), followers (e.g., Dansereau et al.,1975; Graen & Scandura, 1987; Graen & Uhi-

Bien, 1995), and various aspects of situations

(Vroom & Jago, 1988) influence leaderbehaviors. Therefore, various leader behaviorsor behavioral patterns can exist in the same

culture, which differs from a completelydeterministic view of culture. In addition, a

number of leadership theorists view culture as akey factor that may moderate the relationshipbetween leader behavior and leadershipeffectiveness (Ayman, 1993; Bass, 1990;Dorfman et al., 1997; Hartog et al., 1999; House,1995; House et al., 1999; Triandis, 1993).Accordingly, we view culture as a moderator.

In the review of Hofstede’s (2001) secondedition, Smith (2002) emphasized a need for amore fine-grained understanding of what goeson within cultures. Culture at a society level caninfluence regional culture or organizationalculture. However, there can be variations acrossregions within the same society or country.Vandello and Cohen (1999) found that withinthe United States, the South shows relativelymore collectivism, the Great Plains andMountain West shows more individualism, andHawaii shows the most collectivism. In a similar

way, although culture at a society level mayinfluence organizational culture, organizationalculture may differ among organizations withinthe same country. For example, some

organizations may emphasize a value ofindividualistic culture (e.g., equal exchange), butother organizations may emphasize a value ofcollectivist culture (e.g., membership). In theformer organizations, if a leader behaves towardsubordinates on a membership basis,subordinates will not accept the leader’sbehavioral pattern as legitimate. On the otherhand, if a leader behaves toward subordinates ona one-to-one or dyadic basis, subordinates willaccept the leader’s behavioral pattern as

legitimate. Therefore, we can hypothesize thatorganizational culture may also moderate the

relationship between a leader’s behavioral

pattern and acceptance of the leader (i.e.,interdependence between leaders and followers).Clearly, this hypothesis needs empirical testingin the future.

In this paper, we considered only three

leadership approaches (i.e., ALS, VDL, and IL)because our focus is on interdependencebetween leaders and followers. However, otherleadership approaches can be considered. Forexample, Yammarino et al. (1997) suggested



89

that a transformational leader may behavetoward followers on the basis of (1) his or herown style (individual differences), (2) thedifferentiation of in- and out-group, or (3) one-to-one dyadic basis. Therefore, followers indifferent cultures may accept theirtransformational leaders depending on a leader’sbehavioral pattern - i.e., ALS, VDL, or IL.

Practical ImplicationsThe model presented in this paper has some

implications for practice. A leader in a specificculture may behave toward subordinates in thesame group on the basis of one of the three

approaches (i.e., ALS, VDL, and IL). However,in a specific culture, subordinates accept onlyone of the three different approaches of leaderbehaviors. For example, in the cultures of VIand HI (e.g., Sweden, Australia, and UnitedStates) people value equity. So, a leader’s one-on-one dyadic exchange based behavior tends tobe accepted, but a leader’s homogeneous andheterogeneous based behavior tends not to beaccepted. In other words, if a leader treats

subordinates all the same, subordinates who

outperform view his or her equal treatment asunfair because they believe that they should notbe equally treated. In a similar way, if a leaderdivides subordinates in the same group into in-and out-group members and differentially treatsin- and out-group members, subordinates whoperform well but are treated as out-groupmembers because of other members view his orher differential treatment as unfair.

In the culture of HC (e.g., Japan), peoplevalue equal treatment. In other words, people inthe same group should be equally treated on thebasis of membership. Accordingly, a leader’sdifferential treatments of subordinates in thesame group tend not to be accepted. In a similarway, if a leader behaves towards subordinates inthe same group on a one-on-one dyadicexchange basis, subordinates tend not to acceptthe leader, because they have a belief of equaltreatment.

In the culture of VC (e.g., Korea and

China), a leader’s equal treatment tends not to beaccepted. For example, in Korea people valueseniority and so believe that seniors should bebetter treated than juniors. Therefore, a leader’sequal treatment tends not to be accepted. In asimilar way, a leader’s one-on-one dyadic

exchange based behavior tends not to be

accepted, because juniors who outperform willbe treated better than seniors.

We have proposed that a leader’sbehavioral pattern is accepted or unaccepted indifferent cultural contexts. For example, in theculture of HI and VI, a leader’s one-on-one

dyadic based behavioral pattern is accepted. Inthe culture of HC, a leader’s homogeneousbehavioral pattern is accepted, whereas a

leader’s heterogeneous behavioral pattern is

accepted in the culture of VC. These

propositions imply that in the culture of HI andVI, a leader’s heterogeneous and homogeneousbehavioral patterns are not accepted and so

unethical to subordinates. Similarly, in theculture of HC, a leader’s dyadic and

heterogeneous behavioral patterns and a leader’s sdyadic and homogeneous behavioral patterns inthe culture of VC are not accepted and so

unethical to subordinates. Therefore, leadersshould acknowledge that their behaviors becomeunethical to subordinates and show behavioral

patterns that are acceptable and ethical to

subordinates.

Implications for ResearchTesting under what context (culture) a

leader is accepted or not is complex. However,we defined acceptance of the leader as

interdependence between leaders and followersand lack of acceptance as lack of

interdependence between leaders and followers.Therefore, to test the propositions presented inthis paper, the degree of interdependence needsto be assessed. One way to test for

interdependence or lack of interdependence iswith within and between analysis (WABA,Dansereau et al., 1984) - i.e., by assessing thevariation of within-entities (e.g., groups) andbetween-entities (Dansereau & Yammarino,2000; Schriesheim et al, 1995; Schriesheim et al.,2000; Yammarino, 1998). That is, if within-

groups variation is significantly greater than

between-groups variation, interdependencebetween leaders and followers exists. Likewise,if between-groups variation is significantlygreater than within-groups variation,interdependence between leaders and followersalso exists. In a similar way, if between-dyadsvariation is significantly greater than within-

dyads variation, interdependence between



90

leaders and followers exists. For example, in aHC culture (e.g., Japan), if between-groupsvariation of leader behavior is significantlygreater than within-groups variation, leaders andfollowers are interdependent - i.e., followers

accept leaders.In conclusion, we have proposed a multiple

level theory of leadership. We hope that byconsidering a leader’s behavioral patterns,acceptance of leader, and cultural patterns,future research will test whether the propositionsin this paper hold. If not, we hope that the ideasin this paper are a first step in attempting todevelop a multiple level theory of leadership thatspecifies the relationship between leaderbehavior and acceptance of the leader at specificlevels in different cultures.

References

Ayman, R. (1993). Leadership perception: The roleof gender and culture. In M. M. Chemers & R.

Ayman (Eds.), Leadership theory and research(pp. 137-166). San Diego: Academic Press.

Ayman, R,. Chemers, M. M., & Fiedler, F. (1995).The contingency model of leadershipeffectiveness: Its levels of analysis. LeadershipQuarterly, 6, 147-167.

Bass, B. M. (1990). Bass and Stogdill’s handbook ofleadership: Theory, research, and managerialapplication (3rd ed.). New York: Free Press.

Chen, C. C., Meindl, J. R., & Hunt, R. G. (1997).Testing the effects of horizontal and verticalcollectivism: A study of rewards allocation

preferences in China. Journal of Cross-CulturalPsychology, 28, 23-43.

Crouch, A., & Yetton, P. (1988). Manager-subordinate dyads: Relationships among tasksand social contact, manager friendliness andsubordinate performance in managementgroups. Organizational Behavior and HumanDecision Processes, 41, 65-82.

Dansereau, F. (1995). A dyadic approach to

leadership: Creating and nurturing the approachunder fire. Leadership Quarterly, 6, 479-490.

Dansereau, F., Alutto, J. A., & Yammarino, F. J.

(1984). Theory testing in organizationalbehavior: The varient paradigm. EnglewoodCliffs, NJ: Prentice Hall.

Dansereau, F., Graen, G. B., & Haga, W. J. (1975). Avertical dyad linkage approach to leadership informal organizations. Organizational Behaviorand Human Performance, 13, 46-78.

Dansereau, F., & Yammarino, F. J. (2000). Withinand between analysis: The varient paradigm as

an underlying approach to theory building andtesting. In K. J. Klein & S.W. J. Kozlowski

(Eds.), Multilevel theory, research, and methodsin organizations: Foundations, extensions, andnew directions (pp. 425-466). San Francisco:Jossey-Bass.

Dansereau, F., Yammarino, F. J., Markham, S. E.,Alutto, J. A., Newman, J., Dumas, M.,Nachman, S., Naughton, T. Kim, K., Al-Kelabi,S., Lee, S., & Keller, T. (1995). Individualizedleadership: A new multiple level approach.Leadership Quarterly, 6, 413-450.

Dienesch, R. M., & Liden, R. C. (1986). Leader-member exchange model of leadership: Acritique and further development. Academy ofManagement Review, 11, 618-634.

Dorfman, P. W., Howell, J. P., Lee, J. K., Tate, U., &

Bautista, A. (1997). Leadership in Western andAsian countries: Commonalities and differencesin effective leadership processes across cultures.Leadership Quarterly, 8, 233-274.

Eden, D. (1990). Pygmalion without interpersonalcontrast effects: Whole groups gain from raisingmanager expectations. Journal of AppliedPsychology, 75, 394-398.

Erez, M., & Earley, P. C. (1993). Culture, self-

identity, and work. New York: Oxford

University Press.Fiedler, F. E. (1967). A theory of leadership

effectiveness. New York: McGraw-Hill.

Fleishman, E. A. (1957). A leader behavior

description for industry. In R. M. Stogdill & A.

E. Coons (Eds.), Leader behavior: Its

description and measurement. Columbus: OhioState University, Bureau of Business Research.

Fleishman, E. A. (1998). Consideration and structure:Another look at their role in leadership research.In F. Dansereau & F.J. Yammarino (Eds.),Leadership: The multiple-level approaches (pp.51-60). Stamford, CT: JAI Press, 1998.

Graen, G. B., Liden, R. C., & Hoel, W. (1982). Roleof leadership in the employee withdrawal

process. Journal of Applied Psychology, 67,868-872.

Graen, G. B., & Scandura, T. A. (1987). Toward apsychology of dyadic organizing. In L. L.

Cummings & B. M. Staw (Eds.), Research inorganizational behavior (pp. 175-208).Greenwich, CT: JAI Press.

Graen, G. B., & Uhl-Bien, M. (1985). Relationship-based approach to leadership: Development ofleader-member exchange (LMX) theory of

leadership over 25 years: Applying a multi-levelmulti-domain perspective. LeadershipQuarterly, 6, 219-247.

Hartog, D. N. D., House, R. J., Hanges, P. J., Ruiz-Quintanilla, S. A., et al. (1999). Culture specific



91

and cross-culturally generalizable implicitleadership theories: Are attributes ofcharismatic/transformational leadershipuniversally endorsed? Leadership Quarterly, 10,219-256.

Hofstede, G. (1980). Culture’s consequences.Beverly Hills. CA: Sage.

Hofstede, G. (2001). Cultures’ consequences:

Comparing, values, behaviours, institutions, andorganisations across nations. Thousand Oaks,CA: Sage.

House, R. J. (1995). Leadership in the 21st century: A

speculative enquiry. In Howard, A. (Ed.), Thechanging nature of work. SanFrancisco: JosseyBass.

House, R. J., Hanges, P. J., Ruiz-Quintanilla, S. A.,Dorfman, P. W., Javidan, M., Dickson, M.,Gupta, V., et al. (1999). Cultural influences onleadership and organization: Project GLOBE. InW. H. Mobley, M. J. Gessner, & V. Arnold

(Eds.), Advances in Global Leadership (pp.171-233). Stamford, CT: JAI Press.

Kagitcibasi, C. (1997). Individualism andcollectivism. In J. W. Berry, M. H. Segall, & C.

Kagitcibasi (Eds.), Handbook of cross-culturalpsychology (2nd, pp. 1-50). Boston: Allyn &Bacon.

Kagitcibasi, C., & Berry, J. W. (1989). Cross-culturalpsychology: Current research and trends.

Annual Review of Psychology, 40, 493-531.Kerr, S., & Jermier, J. (1978). Substitutes for

leadership: Their meaning and measurement.Organizational Behavior and Human

Performance, 22, 374-403.Kim, U., Triandis, H. C., Kagitcibasi, C., Choi, S., &

Yoon, G. (1994). Individualism andcollectivism: Theory, method, and applications.Newbury Park, CA: Sage.

Liden, R. C., & Maslyn, J. M. (1998).Multidimensionality of leader-member

exchange: An empirical assessment throughscale development. Journal of Management, 24,43-72.

Likert, R. (1961). New patterns of management. NewYork: McGraw-Hill.

Markus, H. R., & Kitayama, S. (1991). Culture andself: Implications for cognition, emotion andmotivation. Psychological Review, 98, 224-253.

Misumi, J. (1995). The development in Japan of theperformance-maintenance (PM) theory of

leadership. Journal of Social Issues, 51, 213-228.Misumi, J., & Peterson, M. F. (1985). The

performance-maintenance (PM) theory of

leadership: Review of a Japanese research

program. Administrative Science Quarterly, 30,198-223.

Mumford, M. D. (1998). Situations, interactions andfollowers: The case of individualized

leadership. In F. Dansereau & F. J. Yammarino

(Eds.), Leadership: The multiple level

approaches (pp. 421-427). Stamford, CT: JAIPress.

Mumford, M. D., Dansereau, F., & Yammarino. F. J.

(2000). Followers, motivations, and level of

analysis: The case of individualized leadership.Leadership Quarterly, 11, 313-340.

Offermann, L. R., & Hellmann, P. S. (1997).Culture’s consequences for leadership behavior:National values in action. Joumal of CrossCultural Psychology, 28, 342-351.

Pascal, R. T., & Athos, A. G. (1981). The art of

Japanese management. New York: Simon &

Schuster.

Podsakoff, P. M., & MacKenzie, S. B. (1998). Anexamination of substitutes for leadership withina levels-of-analysis framework. In F. Dansereau& F. J. Yammarino (Eds.), Leadership: Themultiple-level approaches (pp. 215-258).Stamford, Connecticut: JAI Press.

Roberts, K. H., Hulin, C. L., & Rousseau, D. M.(1978). Developing an interdisciplinary scienceof organizations. San Francisco: Jossey-Bass.

Rousseau, D. (1985). Issues of level in organizationalresearch: Multilevel and cross-level

perspectives. In L. L. Cummings & B. M. Staw

(Eds.), Research in organizational behavior (pp.1-37). Greenwich, CT: JAI Press.

Scandura, T. A. (1999). Rethinking leader-memberexchange: An organizational justice perspective.Leadership Quarterly, 10, 25-40.

Schein, E. H. (1992). Organizational culture and

leadership (2nd, Ed.). San Francisco: Jossey-Bass.

Schriesheim, C. A., Cogliser, C. C., & Neider, L. L.(1995). Is it trustworthy? A multiple levels ofanalysis reexamination of an Ohio leadershipstudy with implications for future research.

Leadership Quarterly, 6, 111-145.Schriesheim, C. A., Castro, S. L., & Cogliser, C. C.

(1999). Leader-member exchange (LMX)research: A comprehensive review of theory,measurement, and data-analytic practices.Leadership Quarterly, 10, 63-113.

Schriesheim, C. A., Castro, S. L., & Yammarino, F.J. (2000). Investigating contingencies: An

examination of the impact of span of

supervision and upward controllingness andleader-member exchange using traditional andmultivariate within- and between-entities

analysis. Journal of Applied Psychology. 85,659-677.

Schrieshiem, C. A., Neider. L. L.. & Scandura. T. A.

(1998). Delegation and leader-member



92

exchange: Main effects, moderators, andmeasurement issues. Academy of ManagementJournal, 41, 298-318.

Schwartz, S. H. (1990). Individualism-collectivism:Critique and proposed refinements. Journal ofCross-Cultural Psychology, 21, 139-157.

Singelis, T. M., Triandis, H. C., Bhawuk, D., &

Gelfand, M. J. (1995). Horizontal and verticaldimensions of individualism and collectivism:A theoretical and measurement refinement.

Cross-Cultural Research, 29, 240-275.Smith, P. B. (2002). Culture’s consequences:

Something old and something new. HumanRelations, 55, 119-135.

Stogdill, R. M., & Coons, A. E. (1957). Leaderbehavior: Its description and measurement.

Columbus: Ohio State University, Bureau ofBusiness Research.

Triandis, H. C. (1990). Cross-cultural studies of

individualism and collectivism. In J. Berman

(Ed.), Nebraska Symposium on Motivation (pp.41-133). Lincoln: University of Nebraska Press.

Triandis, H. C. (1993). Collectivism andindividualism as cultural syndromes. Cross-Cultural Research, 27, 155-180.

Triandis, H. C. (1994). Culture and social behavior.New York: McGraw-Hill.

Triandis, H. C. (1995). Individualism andcollectivism. Boulder, CO: Westview Press.

Triandis, H. C. (1996). The psychologicalmeasurement of cultural syndromes. AmericanPsychologist, 51, 407-415.

Triandis, H. C., & Gelfand, M. J. (1998). Convergingmeasurement of horizontal and verticalindividualism and collectivism. Journal of

Personality and Social Psychology, 74, 118-128.

Vandello, J. A., & Cohen, D. (1999). Patterns ofindividualism and collectivism across the

United States. Journal of Personality and SocialPsychology, 77, 279-292.

Vroom, V. H., & Jago, A. G. (1988). The newleadership: Managing participation in

organizations. Englewood Cliffs, NJ: Prentice-Hall.

Wayne, S. J., Shore, L. M., & Liden, R. C. (1997).Perceived organizational support and leader-member exchange: A social exchangeperspective. Academy of Management Journal,40,82-111.

Weick, K. E. (1978). The social psychology of

organizing. Reading, Mass.: Addison-Wesley.Yammarino, F. J. (1998). Multivariate aspects of the

varient/WABA approach: A discussion and

leadership illustration. Leadership Quarterly, 9,203-227.

Yammarino, F. J., Dubinsky, A. J., Comer, L. B., &

Jolson, M. A. (1997). Women and

transformational and contingent reward

leadership: A multiple-levels-of-analysisperspective. Academy of Management Journal,40, 205-222.

Yukl, G. (1998). Leadership in organizations (4th ed.).Englewood Cliffs, NK: Prentice-Hall.

Kyoungsu Kim is Associate Professor of

Organization in the College of Business

Administration, Chonnam National University.His major fields of interest are culture and

leadership at multiple-levels of analysis. Hisresearch focuses on charismatic leadership,organizational structure, role, culture, and

multiple-levels of analysis. Mailing Address:300 Yongbong-dong, Buk-Ku, College ofBusiness Administration, Chonnam National

University, Kwangju, South Korea. [E-mail:[email protected]]

Fred Dansereau is Professor of Organization andHuman Resources in the School of Managementat the State University of New York at Buffalo.His research interests are in the areas of

leadership and managing at the individual,dyad, group, and collective levels of analysis.Along with others, he has developed a

theoretical and empirical approach to theorizingabout and testing theories at multiple levels ofanalysis. [E-mail: [email protected]]

In Sook Kim is Professor of Nursing Administrationin the College of Nursing, Yonsei University.Her major fields of interest are leadership andhospital management. Her research focuses oncontingency theory of leadership, nursinginformatics, and organizational culture.[e-mail :[email protected]]

Kong Soo Kim is Associate Professor of

Organization in the College of Business

Administration, Kunsan National University.His major fields of interest are motivation andleadership. His research focuses on individualand organizational learning and knowledgemanagement. [e-mail: [email protected]]



a multiple-level theory leadership: impact - dr. hatfield multiple level... · 2015. 7. 24. · a...

Documents